Marvin R Natowicz ^1,2* Margaret L Bauman ³ Stephen M Edelson ⁴

• ¹ Pathology & Laboratory Medicine, Genomic Medicine, Neurology and Pediatrics Institutes, Cleveland Clinic, Cleveland, Georgia, United States
• ² Cleveland Clinic Lerner College of Medicine, School of Medicine, Case Western Reserve University, Cleveland, Ohio, United States
• ³ Department of Anatomy and Neurobiology, Chobanian & Avedisian School of Medicine, Boston University, Boston, Massachusetts, United States
• ⁴ Autism Research Institute, San Diego, California, United States

Autism Spectrum Disorder (hereafter, autism or ASD) is a common and complex neurodevelopmental condition that affects about 1% of the population worldwide (1,2).As currently defined, the core features of autism include persistent deficits in social communication and social interaction in multiple settings accompanied by restricted, repetitive patterns of behavior, interests or activities (3). There is, however, substantial variability in the clinical presentations and natural histories of affected individuals (1,2,4). In most instances the etiological basis for an individual's autism is not determined, the underlying biological mechanisms poorly understood, and clinical care often suboptimal (1,2,4). Gaining a more complete understanding of the processes affecting brain biology in autistic persons is important in meeting one of the major global public health challenges of our time: the development of effective care for autistic persons (4). This essay presents the case for ASD postmortem brain research and the need for ASD brain donations. Why is studying human brain tissue important in autism science? There are many established and useful approaches in learning about the causes and associated neurobiological mechanisms related to ASD, many of which do not require an analysis of human brain tissue. Some of these include various types of epidemiological investigations, clinical cohort analyses, studies of the genetics or biochemistries of peripheral human tissues such as blood or urine, and neuroimaging studies of children and adults. Some of the central lessons learned from these types of studies include knowledge that ASD is a common and both etiologically and clinically highly heterogeneous condition, that autism has high heritability and that there are several neuroradiological findings that are more commonly noted in autistic persons (1,2). In addition, multiple investigations have shown that many rare, highly penetrant genetic variants associated with autism involve such biological processes as chromatin remodeling and synapse organization and function (5). It is therefore important to ask: Why is the procurement of autistic brain tissue of such importance for the advancement of knowledge about autism? While most persons with autism have one or more co-occurring medical conditions that involve tissues outside of the brain (1,2,4), the core involved organ system related to ASD is the brain. It is crucial, then, to understand how, where and when the brain is affected. A major challenge in accomplishing this goal is that many important characteristics and processes of brain biology can only be understood through the direct investigation of brain tissue obtained from ASD individuals. What types of studies can be carried out using human brain tissue and what types of information can be learned from those studies? Many different types of studies can be carried out using human brain tissue, all of which have the potential to provide valuable information about ASD. Some of the major types of studies that have been carried out include histological analyses, gene expression studies and various types of biochemical analyses including measurements of proteins or lipids or other metabolites. Each of the major types of studies that have been carried out on postmortem ASD brain is briefly reviewed below. Histological and Histochemical Analyses Analyzing brain tissue from individuals with ASD offers the opportunity to examine ASD-related biological changes associated with neuronal and glial cellular organization, synaptic connectivity and neurotransmitter systems. By studying brain tissue, researchers can observe how neurons and other cell types are arranged and connected, revealing potential abnormalities in the neural networks that are critical for communication, behavior, and cognition. Analyses of neurotransmitter systems in brain tissue can provide insights into potential chemical imbalances that may contribute to the symptoms of autism, thus offering possible targets for pharmacological interventions. In addition, research involving the study of brain tissue can also provide insights into brain plasticity, the brain's ability to change in response to experiences. This can be especially relevant for ASD individuals where an understanding of potential changes that might be part of development and/or in response to medical and therapeutic interventions may provide insight into the brain processes of ASD and suggest specific therapeutic approaches. Histological/histochemical analyses of ASD postmortem brain have yielded important but sometimes varied or inconsistent results. These differences are likely attributable to variations in the brain regions evaluated in different studies, differences in methodological approaches, differences in underlying etiology and variation in the ages and sex of the donors, as well as differences in co-occurring conditions, treatments and causes of death (6,7). Histological abnormalities have been noted in different regions of the cerebral cortex (eg, prefrontal and frontal cortex, fusiform gyrus, frontoinsular and cingulate cortex, hippocampus) (6,7). Non-cortical regions consistently altered in ASD include the amygdala and cerebellum (6,7). Abnormalities that have been observed include differences in neuronal size, number and density and in the disposition of their axons and dendritic spines, as well as histological changes in non-neuronal cells and in myelination (6,7). A substantial subset of ASD brains show evidence of neuroinflammation (8). Many of the observed histological differences relate to brain regions involved in motor control, cognition and emotional regulation. Gene Expression Analyses Additional information regarding the biology of autism that has been obtained through analyses of postmortem brain tissue has included studies of ASD gene expression changes, the up-or down-regulation of different genes in the autistic brain compared to control brain tissue. The regulation of gene expression is complex and requires the normal function of different steps from the initial transcription of the information present in DNA to the final production and localization of the end product RNAs, messenger RNAs and non-coding RNAs. Dysregulation of gene expression can be due to an inherited or de novo genetic variant or consequent to a non-genetic process that alters one or more steps of the gene expression process. Non-genetic occurrences that can impact brain gene expression include changes in the prenatal or postnatal brain cellular environment due to altered levels or new exposures to hormones, nutritional changes, infections, toxins and still other factors and are sometimes mediated by epigenetic processes. Regardless of the cause of gene dysregulation, changes in gene expression can be identified by measurement of the levels of many mRNA and non-coding RNA species (transcriptomic analysis) and, in cases of epigenetic dysregulation, specific epigenetic signatures, such as particular gene methylation patterns, can be identified. Many changes in ASD gene expression patterns have been identified and some convergence among dysregulated genes and pathways revealed, including in ASD brain (9-11). Most ASD brain gene expression studies have involved analyses of tissue samples consisting of numerous brain cells, sometimes referred to as bulk gene expression studies, an approach lacking cell-specific resolution. Use of newer methods to interrogate single brain cells or single cell nuclei has enabled gene expression studies to be carried out on specific cell types from different regions of postmortem brain. These powerful approaches, in turn, have disclosed cell-specific information that has confirmed and extended the gene expression analyses of larger brain tissue samples (12-18). Information derived from detailed gene expression, other genetic and epigenetic and biochemical (below) analyses on single cells, when integrated with information from cells across all brain regions and across ages and phenotypes, will facilitate markedly increased understandings of underlying ASD brain biology (19,20). Like the abovementioned histological studies, research on changes in ASD brain gene expression rely on investigations using donated ASD postmortem brain tissue. Biochemical Analyses In addition to histological and gene expression studies of ASD postmortem brain tissue, there have been different types of biochemical studies ranging from analyses of specific metabolites or families of metabolites to analyses of one or more proteins in autistic brain. It is beyond the scope of this essay to describe the large number of studies that involve analyses of a single or several metabolites or proteins whereas to date there are still only a few large-scale metabolic (metabolomic) (21,22) or large scale protein (proteomic) (23-25) analyses of ASD brain. The power of this level of experimental use of brain tissue is that the molecules of study - various proteins or metabolites - are the 'agents' of altered gene expression and thereby 'closer' to the phenotype of altered brain function in autism. The diverse investigations of autism using autistic brainhistological, gene expression and biochemicalhave yielded important insights regarding the biology of autistic brain, most of which could not have been achieved without the availability of ASD postmortem brain specimens. An early study of ASD brain is an historically important example of this. The first substantive investigation of postmortem ASD brain, a histological analysis of a single adult brain published in 1985, revealed distinctive regional histological abnormalities and, importantly, surmised that at least in that individual those brain abnormalities had to have occurred during prenatal development (26). This, in turn, fundamentally challenged the then commonly held hypothesis of autism as a consequence of poor parenting. What are the major challenges for scientists needing brain tissue for ASD research? Researchers who use ASD brain for their scientific work are faced with important challenges. One significant challenge is the limited number of ASD brains that are available for investigative work, as well as the related issue that some available specimens are poorly characterized or that the quality of the tissue for investigative purposes is compromised. An additional common limitation is that detailed clinical and other relevant ancillary information about the individuals whose brains have been donated are often incomplete and this, in turn, limits the power and, sometimes, even the use of some donated brains for some important research projects. Currently, the NIH NeuroBioBank network (27) has brain specimens from >3200 Alzheimer disease subjects, >400 amyotrophic lateral sclerosis disease subjects, >700 Huntington disease subjects, >1600 multiple sclerosis disease subjects, >1000 subjects with schizophrenia and >1100 Parkinson disease subjects, but there are only 106 brain specimens from ASD subjects (28). Similarly, the total number of fixed or frozen brain tissues available through the Autism BrainNet network (29) includes 204 specimens, although the number of individual subjects is less than 204 as the brain tissue from some subjects was divided into both fixed and frozen hemispheres (30). While each of these neurological disorders has much personal and societal clinical significance and is in need of continued research attention, the disparity in the number of ASD postmortem brains that are available for research compared to the number of banked brains from many other conditions is striking, especially when the number of ASD brains is considered relative to the number of available brains from other conditions of approximately similar or even lesser population prevalence (31,32). In 2021, for example, the age-standardized population prevalence of ASD was 783.3 individuals per 100,000 people and the population prevalence of Alzheimer disease and other dementias, Parkinson disease, multiple sclerosis and motor neuron disease were, respectively, 694/100,000, 138.6/100,000, 22.2/100,000 and 3.3/100,000 (32) yet, as noted above, each of these other neurological conditions has considerably greater numbers of postmortem brains available for research at the NIH NeuroBioBank. What is the Process for Donating to a Brain Tissue Bank? Parents and family members should discuss the possibility of donating their loved one's brain to a tissue bank for research use. This decision can be difficult and may be influenced by personal, religious, and/or cultural beliefs. If the decision to donate the loved one's brain to a tissue bank for research purposes has been made in advance of the individual's death, the family can contact a brain bank in advance. Contact with the brain bank before a donor's death is important, when possible. The passing of a loved one is usually a stressful time and having to locate and contact a tissue bank and then provide necessary information can make the situation even more challenging. Autism BrainNet (www.AutismBrainNet.org) consists of a network of three U.S. sites and two international partnerships. There is a 24-hour hotline number (+1-877-333-0999) that a family member can call to initiate the donation process. The answering staff will then walk the family through the process of brain donation. This process involves a determination if the donor meets inclusion and exclusion criteria and, if so, a consent process is undertaken to authorize the research use of the brain and the acquisition of medical records. A clinical director then facilitates the process for obtaining the brain donation in the shortest time by working with a local coroner or regional collaborating brain bank (29). In some instances the demise of a loved one is unanticipated, sometimes due to an accident or a new illness, and there may not have been prior consideration of the possibility of a brain donation. In this situation, donation of the brain can still occur. The family can contact the above 24-hour hotline and the donation process can proceed as described. In addition, pathologists, coroners and medical examiners can also contact Autism BrainNet for assistance with ASD brain donation in instances when potential donors are first encountered at autopsy (33). There are other high quality brain banks available for families considering brain donation. The National Institute of Health NeuroBioBank, accessible through the Brain Donor Project (www.BrainDonorProject.org; 513-393-7878), oversees a network of brain banks across six U.S. locations. These sites collect brain tissue from individuals with various neurological, neuropsychiatric and neurodevelopmental disorders. The participating institutions include the University of Maryland Brain and Tissue Bank (800-847-1539), the University of Miami Brain Endowment Bank (800-UM-BRAIN), the Brain Tissue Donation Program at the University of Pittsburgh (513-393-7878), the Harvard Brain Tissue resource Center (800-BRAIN-Bank), the Mount Sinai Brain Bank (212-807-5541), and the Human Brain and Spinal Fluid Resource Center (310-268-3330). Both the Autism BrainNet and the NIH NeuroBioBank networks encourage preregistration and have user friendly web-based and phone-based systems to accomplish this. The Autism BrainNet network is focused on advancing autism brain research; the NIH NeuroBioBank network is a resource for research on many types of brain disorders. Both networks, as well as other brain banks (34-36), accept donations of neurotypical brains. The importance of the latter also needs special mention; donated neurotypical brains serve as critically important controls in scientific investigation (37). High quality postmortem-based brain research requires close matching of cases and controls with respect to all key parameters, typically including matching for age, sex, postmortem interval and region of brain studied. Additionally, there often are other crucial inclusion/exclusion criteria such as the absence of certain conditions (e.g., brain trauma, brain hemorrhage, primary or metastatic brain tumor, recent stroke or seizure, use of certain medications, etc.). Having a large pool of control brains increases the likelihood that control brain samples are not exhausted and that cases and controls can be appropriately matched for all essential variables. Overall, the donation of a brain for research purposes is a complex process, from the procurement of the brain to its ultimate use in research investigations. Each step of the process requires forethought and appropriate implementation, ranging from the respectful interaction with the family and collection of needed medical information to time-sensitive logistical issues regarding the timely handling and processing of the brain to the dispensing of tissue for needed investigation. Finally, we recognize that not all families or pathologists that may want to be involved in ASD brain donation have close geographic proximity to a center specializing in this effort. The brain banks of the NIH are not able to accept brains from outside of the United States. Autism BrainNet has two collaborating centers outside of the United States: the Douglas-Bell Canada Brain Bank (douglasbrainbank.ca; 514-761-6131ext.3496), a McGill University affiliate, and the Oxford Brain Bank in the United Kingdom (brainbank@ndcn.ox.ac.uk). For those situations where a brain donation from an ASD donor is desired and the deceased individual is not in a location with timely access to one of the described networks, the family or pathologist might consider an online search for brain donation programs in that country or region. Conclusion The collection and analysis of human brain tissue is indispensable for advancing the understanding of ASD. Current technologies are unable to study living human brain at levels of resolution that are important in understanding the cellular and molecular alterations that are present in ASD brain. Thus, much vital information about the biological basis of autism can only be learned through investigation of postmortem brain. Insights gained from studying postmortem brain tissues result in an improved understanding of the cellular and molecular basis of the neurobiology of autism. This, in turn, will be helpful in enabling the development of innovative therapeutic approaches that hold the promise of transforming lives.